Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method of associating device activity on a first device to a broadcast programme broadcast on a second device, the first device being different than the second device, the method comprising: at the second device: receiving the broadcast programme; and displaying the broadcast programme; at the first device: monitoring, via a client software module running on the first device, an activity stream of one or both of the first device's active and inactive states; and transmitting the activity stream from the first device; at a server: receiving the broadcast programme displayed on the second device; generating a model for the broadcast programme; identifying portions of content of the broadcast programme and portions of breaks in the content; receiving the activity stream of the first device from the first device; comparing the activity stream of the first device to the model for the broadcast programme and identifying a correlation between the activity stream and the model for the broadcast program; and inferring a user of the first device as watching the broadcast programme on the second device based on a level of the correlation.
This invention relates to a system for associating device activity on a first device with a broadcast program displayed on a second device. The technology addresses the challenge of determining whether a user is engaged with a broadcast program when their activity is observed on a separate device, such as a smartphone or tablet, rather than the device displaying the content. The method involves multiple components working together. The second device receives and displays the broadcast program. Meanwhile, the first device runs a client software module that monitors its activity stream, including both active and inactive states. This activity data is transmitted to a server. The server also receives the broadcast program displayed on the second device and generates a model of the program, identifying segments of content and breaks. The server then compares the activity stream from the first device to the program model to detect correlations. If a strong correlation is found, the system infers that the user of the first device is likely watching the broadcast program on the second device, even though the activity was observed on a different device. This approach enables more accurate audience measurement and engagement tracking across multiple devices.
2. The method of claim 1 , wherein the model is a user behaviour model, the method further comprising: updating the user behaviour model in dependence on the comparing step.
A system and method for analyzing and updating user behavior models to improve decision-making processes. The technology addresses the challenge of accurately predicting user actions in dynamic environments where behavior patterns evolve over time. The method involves generating a user behavior model that represents expected user actions based on historical data or predefined rules. This model is then compared against observed user actions in real-time or near-real-time to identify discrepancies. When a mismatch is detected, the user behavior model is updated to incorporate new data, ensuring the model remains accurate and relevant. The updating process may involve adjusting model parameters, refining algorithms, or integrating additional data sources to reflect changing user behaviors. This adaptive approach enhances the reliability of predictions, enabling more effective decision-making in applications such as personalized recommendations, fraud detection, or user experience optimization. The system ensures continuous improvement by dynamically refining the model in response to observed behavior, reducing errors and improving system performance over time.
3. The method of claim 1 , wherein the comparing step compares activity indicated in the activity stream with periods of content and periods of break in the model.
This invention relates to analyzing user activity data to improve productivity or engagement. The problem addressed is the lack of effective tools to correlate user activity patterns with structured work or rest periods, which can help optimize workflows or identify inefficiencies. The method involves processing an activity stream, which may include timestamps, actions, or other indicators of user engagement. This activity data is compared against a predefined model that defines periods of content (e.g., focused work) and periods of breaks (e.g., rest or idle time). The comparison identifies mismatches or deviations between the user's actual activity and the ideal model, allowing for adjustments to improve alignment. The model may be based on productivity research, user preferences, or historical data. The comparison step evaluates whether the user's activity aligns with the expected content and break periods in the model. For example, if the model suggests a 25-minute work session followed by a 5-minute break, the method checks if the user's activity matches this pattern. Discrepancies may trigger alerts, recommendations, or automated adjustments to the user's schedule or environment. The invention may be applied in productivity software, workplace monitoring systems, or health and wellness applications to enhance efficiency or well-being. The core innovation lies in dynamically assessing activity against a structured model to provide actionable insights.
4. The method of claim 3 , wherein high activity during the break periods and minimal to no activity during the content periods infers that the user of the first device is watching the programme on the second device.
This invention relates to determining whether a user is watching a program on a second device while interacting with a first device, such as a smartphone or tablet. The problem addressed is the difficulty in accurately detecting user engagement with content on one device while simultaneously using another device, which is important for advertising, analytics, and user experience optimization. The method involves monitoring user activity on the first device, particularly during breaks in the content being displayed on the second device. High activity on the first device during these break periods, combined with minimal or no activity during the content periods, indicates that the user is likely watching the program on the second device. This inference is based on the assumption that users typically engage with their first device during commercials or pauses but focus on the second device during the main content. The system may also track additional contextual factors, such as the timing of activity spikes, the type of content being displayed, and historical user behavior, to improve the accuracy of the inference. This approach helps distinguish between passive multitasking and active engagement with the second device, providing more reliable insights for content providers and advertisers.
5. The method of claim 1 , further comprising: determining that the user of the first device is watching the broadcast programme on the second device, the comparing step being enabled in dependence thereon.
A system and method for monitoring and analyzing broadcast program viewing behavior involves tracking user interactions with multiple devices to determine engagement with a broadcast program. The method includes detecting when a user of a first device, such as a mobile device, is interacting with a broadcast program, such as by accessing related content or performing program-specific actions. The system then identifies whether the same user is simultaneously watching the broadcast program on a second device, such as a television. If the user is confirmed to be watching the program on the second device, the system enables a comparison step to analyze the user's interactions on the first device in relation to their viewing behavior on the second device. This comparison may involve correlating timestamps, interaction patterns, or other engagement metrics to assess the user's overall engagement with the program across devices. The method aims to provide insights into cross-device viewing habits, improving content personalization, advertising effectiveness, or user experience optimization. The system may also adjust recommendations or notifications based on the detected viewing context.
6. The method of claim 5 , wherein the determining that the user of the first device is watching the broadcast programme on the second device comprises one or more of: collecting a locked status of one of both of the first device and the second device; collecting a screen interaction status of one of both of the first device and the second device; collecting a motion activity status of one of both of the first device and the second device; collecting a screen power status of one of both of the first device and the second device; collecting a processor unit state of one of both of the first device and the second device; collecting a battery drain status of one of both of the first device and the second device; and collecting active application states of one of both of the first device and the second device.
This invention relates to determining whether a user of a first device is watching a broadcast program on a second device by analyzing various status indicators from both devices. The technology addresses the challenge of accurately detecting user engagement with media content across multiple devices, which is important for targeted advertising, content recommendations, and user analytics. The method involves collecting multiple types of status data from the first and second devices to infer whether the user is actively watching the broadcast program. These status indicators include the locked status of either device, screen interaction status, motion activity status, screen power status, processor unit state, battery drain status, and active application states. By analyzing these factors, the system can determine whether the user is likely engaged with the broadcast program on the second device while using the first device. This approach improves the accuracy of user activity tracking by considering multiple device states rather than relying on a single metric. The solution is particularly useful in scenarios where users switch between devices or use one device as a secondary screen while watching content on another.
7. The method of claim 1 , further comprising: storing a usage pattern and time stamps of one of both of the first device and the second device; aligning historical usage pattern data based on similar time frames; normalizing current usage pattern data via aligned historical usage pattern data; and utilizing the normalized pattern as the activity stream of the user in the comparing step.
This invention relates to analyzing and comparing usage patterns of electronic devices to determine user activity. The problem addressed is the difficulty in accurately identifying and comparing user behavior across different devices due to variations in usage patterns and time frames. The solution involves collecting and processing usage data from at least two devices to generate a normalized activity stream for comparison. The method includes storing usage patterns and timestamps from one or both devices, aligning historical usage data based on similar time frames, and normalizing current usage data using the aligned historical data. This normalization process ensures that variations in usage patterns are accounted for, allowing for a more accurate comparison of user activity. The normalized pattern is then used to compare the activity streams of the devices, enabling better detection of user behavior and interactions. By aligning and normalizing usage data, the invention improves the accuracy of activity analysis, making it easier to track user behavior across multiple devices. This approach is particularly useful in applications such as user authentication, activity monitoring, and behavioral analysis, where consistent and reliable data is essential. The method ensures that temporal and usage variations do not distort the comparison results, providing a more reliable assessment of user activity.
8. The method of claim 1 , further comprising: restricting results of the comparing step to one or both of popular broadcast programmes and broadcast programmes for which the user has an affinity.
This invention relates to a system for comparing broadcast program data to determine similarities between programs. The method involves analyzing program metadata, such as titles, descriptions, and genres, to identify matching or similar programs. The comparison results can be filtered to focus on either popular broadcast programs or programs that align with a user's preferences, ensuring the output is relevant and tailored to the user's interests. By restricting the comparison to these subsets, the system enhances the efficiency and accuracy of program recommendations or comparisons, avoiding irrelevant or low-engagement content. The filtering step ensures that only the most relevant programs are considered, improving the user experience by prioritizing content that is either widely watched or personally appealing. This approach is particularly useful in recommendation systems, content discovery platforms, or broadcast scheduling tools where personalized or trending content is prioritized. The method leverages metadata analysis to refine search results, making it easier for users to find programs that match their tastes or current trends.
9. The method of claim 1 , further comprising: notifying the inferred broadcast programme to the user of the first device, and receiving notification as to whether the inference is correct.
A system and method for inferring and validating broadcast program content involves analyzing audio or video signals from a first device to identify a broadcast program being played. The system processes the signals using machine learning models or pattern recognition techniques to determine the program's identity, such as its title, genre, or source. Once inferred, the system notifies the user of the first device with the inferred program details. The user can then provide feedback by confirming whether the inference is correct or not. This feedback loop helps improve the accuracy of future inferences by refining the underlying models. The method may also involve comparing the inferred program with a database of known programs or using metadata from the broadcast signal to enhance accuracy. The system can be applied in smart TVs, streaming devices, or other media playback systems to automatically detect and validate broadcast content for personalized recommendations, parental controls, or usage analytics. The feedback mechanism ensures continuous improvement in program recognition accuracy.
10. A computer device comprising non-transitory storage media for storing computer program code which, when executed, performs the method of claim 1 .
A computer device is designed to address the challenge of efficiently executing computational tasks by leveraging optimized software instructions. The device includes non-transitory storage media that holds computer program code. When executed, this code performs a method that involves processing input data through a series of computational steps to generate an output. The method may include receiving input data, applying one or more algorithms to transform the data, and producing a result based on the transformations. The device is particularly useful in applications requiring high-performance computing, such as data analysis, machine learning, or real-time processing, where efficiency and accuracy are critical. The storage media ensures that the program code remains persistent and accessible for repeated execution, while the computational steps are optimized to minimize resource usage and maximize throughput. This approach enhances the device's ability to handle complex tasks with improved speed and reliability, making it suitable for various technical and industrial applications.
11. A server for associating activity on a first device to a broadcast programme broadcast on a second device, the first and second devices being different devices, the server comprising: a receiver for receiving the broadcast programme broadcast on the second device; a module for generating a model of the broadcast programme, the model identifying portions of content and portions of breaks in the content; a second receiver for receiving an activity stream of the first device, the activity stream denoting one or both of the first device's active and inactive states; a comparator for comparing the activity stream with the model and for identifying a correlation between the activity stream and the model; and a module for inferring a user of the first device as watching the programme on the second device in dependence on the correlation.
This invention relates to a server system that associates user activity on a first device with a broadcast program being displayed on a second device. The technology addresses the challenge of determining whether a user is engaged with a broadcast program on one device while interacting with another device, such as a smartphone or tablet. The system operates by analyzing the broadcast program and user activity to infer viewing behavior. The server receives the broadcast program from the second device and generates a model that segments the program into content portions and ad breaks. Simultaneously, the server collects an activity stream from the first device, which tracks the device's active and inactive states. By comparing the activity stream with the program model, the server identifies correlations between user activity and the broadcast content. For example, if the first device becomes inactive during a program segment and active during ad breaks, the system infers that the user is likely watching the program on the second device. This correlation-based approach enables accurate attribution of user engagement across multiple devices without requiring direct input from the user. The system enhances targeted advertising and content recommendations by improving the understanding of cross-device media consumption patterns.
12. The server of claim 11 , wherein the model is a user behaviour model, the server further being configured to update the user behaviour model in dependence on the correlating the activity stream with the model.
This invention relates to a server system that processes user activity data to generate insights or predictions based on user behavior. The system collects an activity stream representing user interactions with a digital platform, such as clicks, navigation patterns, or content consumption. The server includes a model, specifically a user behavior model, which analyzes the activity stream to identify patterns, preferences, or predictive outcomes. The model is dynamically updated based on the correlation between the activity stream and its existing predictions, ensuring continuous refinement of accuracy. The server may also include a data processing module to preprocess the activity stream, such as filtering or normalizing the data, before analysis. Additionally, the system may generate output signals, such as recommendations, alerts, or personalized content, based on the model's analysis. The invention aims to improve the accuracy and adaptability of user behavior modeling in digital environments, enabling more effective personalization or decision-making. The dynamic updating mechanism ensures the model remains relevant as user behavior evolves over time.
13. The server of claim 11 , wherein the server is configured to compare activity indicated in the activity stream with periods of content and periods of break in the model.
This invention relates to a server system for analyzing user activity data, particularly in the context of digital content consumption or productivity tracking. The problem addressed is the need to accurately model and compare user activity patterns against predefined or learned activity models to assess engagement, productivity, or other behavioral metrics. The server receives an activity stream containing timestamps and event data representing user interactions with a system, such as clicks, scrolls, or content consumption events. It processes this stream to identify periods of active engagement (content periods) and inactive or idle periods (breaks). The server then compares these detected activity patterns against a predefined or dynamically generated model that defines expected or optimal activity sequences. The model may include rules, thresholds, or statistical distributions for distinguishing between productive and unproductive periods. The comparison step involves aligning the activity stream data with the model's defined periods to determine deviations, such as excessive breaks, early disengagement, or irregular activity patterns. This analysis can be used for applications like productivity monitoring, content personalization, or user behavior analytics. The system may further adjust the model based on historical data or user-specific trends to improve accuracy over time. The invention aims to provide a scalable, automated way to assess user activity against expected patterns for various applications.
14. The server of claim 13 , wherein the server is configured to infer from high activity during the break periods and minimal to no activity during the content periods that the user is watching the programme.
A system for analyzing user engagement with digital content, particularly video or audio programs, determines whether a user is actively watching or listening based on device activity patterns. The system monitors user interactions, such as keyboard, mouse, or touchscreen inputs, during both content playback and scheduled break periods (e.g., commercials or intermissions). By comparing activity levels between these intervals, the system infers user engagement. Specifically, if high activity is detected during breaks but minimal or no activity occurs during content playback, the system concludes that the user is likely watching or listening to the program. This approach helps distinguish between active engagement and passive or background usage, improving metrics for content providers and advertisers. The system may integrate with media playback devices, streaming platforms, or smart home systems to collect and analyze activity data in real time. The analysis may also account for variations in user behavior, such as brief pauses or multitasking, to enhance accuracy. The solution addresses challenges in measuring true audience engagement, particularly in environments where devices remain active but users may not be fully attentive.
15. The server of claim 11 , wherein the server is further configured to watch broadcast content, and enable a comparison in dependence on.
A system for monitoring and analyzing broadcast content involves a server that receives and processes media streams from various sources. The server includes a content analysis module that extracts metadata, such as timestamps, program identifiers, and audio-visual features, from the broadcast content. This metadata is stored in a database for further processing. The system also includes a comparison module that compares the extracted metadata against reference data to detect similarities, discrepancies, or unauthorized usage of the content. The comparison may involve pattern recognition, fingerprinting, or other analytical techniques to identify matches or deviations. The server can be configured to monitor live broadcasts, recorded content, or archived media. The comparison results are used to generate alerts, reports, or enforcement actions, such as blocking unauthorized transmissions or logging compliance violations. The system may also integrate with external databases or APIs to enhance the accuracy and scope of the comparisons. This technology is particularly useful in media rights management, content protection, and compliance monitoring, ensuring that broadcast content is used in accordance with licensing agreements and regulatory requirements.
16. The server of claim 15 , wherein the server is configured to determine if the user is watching the broadcast content in dependence on one or more of: collecting the device's locked status; collecting the device's screen interaction status; collecting the device's motion activity status; collecting the device's screen power status; collecting the device's processor unit state; collecting the device's batter drain status; and collecting the device's active application states.
This invention relates to a server system for monitoring user engagement with broadcast content, particularly to determine whether a user is actively watching the broadcast. The problem addressed is the need for accurate detection of user attention to broadcast content, which is crucial for advertising, analytics, and content personalization. Traditional methods often rely on simplistic metrics like device activity, which can be unreliable. The server is configured to assess user engagement by analyzing multiple device parameters. These include the device's locked status, screen interaction status, motion activity status, screen power status, processor unit state, battery drain status, and active application states. By evaluating these factors, the server can infer whether the user is actively watching the broadcast content. For example, if the device is unlocked, the screen is on, there is motion activity, and the broadcast application is active, the server can conclude that the user is likely engaged. Conversely, if the device is locked, the screen is off, or the battery is draining rapidly, the server may determine that the user is not watching. This multi-faceted approach improves the accuracy of user engagement detection compared to single-factor methods. The system can be used in various applications, such as targeted advertising, content recommendation, and audience analytics.
17. The server of claim 11 , further comprising: a store for storing a usage pattern and time stamps of the first device, and configured to: align historical usage pattern data based on similar time frames; normalize current usage pattern data via aligned historical usage pattern data; and utilize the normalized pattern as the activity stream of the user in the correlating the activity stream with the model.
This invention relates to a server system for analyzing and correlating user activity patterns with predictive models. The system addresses the challenge of accurately interpreting user behavior by normalizing current usage data against historical patterns to improve model correlation. The server includes a storage component that records usage patterns and timestamps from a first device. The system aligns historical usage data based on similar time frames to establish a baseline. It then normalizes current usage data by comparing it to this aligned historical data, ensuring consistency in the activity stream. This normalized pattern is used to correlate with a predictive model, enhancing the accuracy of behavior analysis. The server also includes a communication interface for receiving and transmitting data, a processor for executing instructions, and a memory for storing executable code. The system may further include a model training module that updates the predictive model based on the correlated activity stream, refining its accuracy over time. Additionally, a user interface module allows for interaction with the system, enabling adjustments to the model or data visualization. By normalizing current usage patterns against historical data, the system improves the reliability of activity stream analysis, making it more effective for applications such as user behavior prediction, anomaly detection, or personalized recommendations.
18. The server of claim 11 , further configured to restrict results of the correlating the activity stream with the model to one or both of popular broadcast programmes and broadcast programmes for which the user has an affinity.
This invention relates to a server system for analyzing user activity data in relation to broadcast media, such as television or streaming content. The problem addressed is the need to filter and prioritize broadcast program recommendations based on user preferences and popularity trends. The server is configured to process an activity stream, which includes user interactions with media content, and correlate this data with a predictive model. The model is trained to identify patterns in user behavior and content attributes. The server further refines the results by restricting them to either popular broadcast programs or programs for which the user has a demonstrated affinity. Affinity may be determined through historical viewing data, explicit user ratings, or inferred preferences. The system aims to improve recommendation accuracy by focusing on content that aligns with both broad trends and individual user tastes. The server may also adjust the filtering criteria dynamically based on real-time data or user feedback. This approach enhances personalization while ensuring recommendations remain relevant and engaging.
19. The server of claim 11 , further configured to notify the user of the first device as to the inferred programme, and to receive notification as to whether the inference is correct.
This invention relates to a server system for inferring and validating user preferences in a multi-device environment. The problem addressed is the difficulty of accurately determining a user's intended media content or program across different devices without requiring explicit input. The system includes a server that monitors user interactions with a first device, such as a smartphone or tablet, to infer the user's likely intended program or content. The server then notifies the user of the inferred program and receives feedback on whether the inference is correct. This feedback loop improves the accuracy of future inferences. The server may also manage synchronization between multiple devices, ensuring that the inferred program is available on a second device, such as a television or smart speaker, for seamless playback. The system may use historical usage data, contextual cues, or direct user input to refine its inferences over time. The goal is to reduce manual selection efforts while maintaining high accuracy in content recommendations.
20. A method for associating activity on a first device to a broadcast programme broadcast on a second device, the first and second devices being different devices, the method comprising: receiving the broadcast programme broadcast on the second device; generating a model of the broadcast programme, the model identifying portions of content and portions of breaks in the content; receiving an activity stream of the first device, the activity stream denoting one or both of an active state of the first device and an inactive state of the first device; comparing the activity stream with the model, and identifying a correlation between the activity stream and the model; and inferring a user of the first device as watching the programme on the second device in dependence on the correlation.
This invention relates to associating user activity on a first device with a broadcast program being displayed on a second device. The problem addressed is determining whether a user is engaged with a broadcast program on one device while interacting with another device, such as a smartphone or tablet. The method involves receiving the broadcast program from the second device and generating a model that segments the program into content portions and ad breaks. An activity stream from the first device is also received, indicating periods of active or inactive use. The activity stream is compared to the program model to identify correlations, such as increased activity during content segments and reduced activity during breaks. Based on these correlations, the system infers that the user is likely watching the broadcast program on the second device while using the first device. This approach enables targeted advertising, audience measurement, or personalized content recommendations by linking cross-device behavior to broadcast content. The method does not require explicit user input or direct device synchronization, relying instead on passive monitoring and pattern recognition.
Unknown
September 15, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.